LibenHailu opened a new issue, #40902:
URL: https://github.com/apache/beam/issues/40902

   ### What happened?
   
   There are two bugs in Beam YAML windowing transforms:
   
   #### 1. `YamlProviders.WindowInto._parse_duration` defines `'d'` (days) as 
12 hours instead of 24 hours
   In `sdks/python/apache_beam/yaml/yaml_provider.py` 
(`YamlProviders.WindowInto._parse_duration`), the time unit multiplier for 
`'d'` is defined as:
   ```python
   time_units = {
       'ms': 0.001, 's': 1, 'm': 60, 'h': 60 * 60, 'd': 60 * 60 * 12
   }
   ```
   Because `60 * 60 * 12` is `43,200` seconds (12 hours) rather than `86,400` 
seconds (24 hours), any `WindowInto` or `windowing` configuration using the 
`'d'` suffix (e.g., `size: 1d`) creates 12-hour windows instead of 24-hour 
windows.
   
   **Steps to Reproduce:**
   ```python
   from apache_beam.yaml import yaml_provider
   
   # Returns 43200.0 instead of 86400.0
   print(yaml_provider.YamlProviders.WindowInto._parse_duration('1d', 'size'))
   ```
   
   ---
   
   #### 2. `_ExtractWindowingInfo` fails to detect duplicate field names in the 
input schema
   In `sdks/python/apache_beam/yaml/yaml_mapping.py` (`_ExtractWindowingInfo`), 
`existing_fields` is obtained via 
`named_fields_from_element_type(pcoll.element_type)`, which returns a list of 
`(field_name, field_type)` tuples (`list[tuple[str, type]]`).
   
   However, the duplicate field check does:
   ```python
   elif field in existing_fields:
     raise ValueError(f'Input schema already has a field named {field}.')
   ```
   Because `field` is a `str` and `existing_fields` contains `(str, type)` 
tuples, `field in existing_fields` always evaluates to `False`. Instead of 
raising `ValueError`, `_ExtractWindowingInfo` appends the duplicate field to 
the output `RowTypeConstraint` (`existing_fields + new_fields`).
   
   **Steps to Reproduce:**
   ```python
   import apache_beam as beam
   from apache_beam.yaml.yaml_transform import YamlTransform
   
   with beam.Pipeline() as p:
     elements = p | beam.Create([beam.Row(timestamp=1, value='a')])
     # Should raise ValueError("Input schema already has a field named 
timestamp.")
     # Instead, produces an invalid RowTypeConstraint with duplicate 
'timestamp' fields
     _ = elements | YamlTransform(
         '''
         type: ExtractWindowingInfo
         config:
           fields: [timestamp]
         ''')
   ```
   
   ### Issue Priority
   
   Priority: 2 (default / most bugs should be filed as P2)
   
   ### Issue Components
   
   - [x] Component: Python SDK
   - [ ] Component: Java SDK
   - [ ] Component: Go SDK
   - [ ] Component: Typescript SDK
   - [ ] Component: IO connector
   - [x] Component: Beam YAML
   - [ ] Component: Beam examples
   - [ ] Component: Beam playground
   - [ ] Component: Beam katas
   - [ ] Component: Website
   - [ ] Component: Infrastructure
   - [ ] Component: Spark Runner
   - [ ] Component: Flink Runner
   - [ ] Component: Prism Runner
   - [ ] Component: Hazelcast Jet Runner
   - [ ] Component: Google Cloud Dataflow Runner
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to